Combined oxygen and glucose sensing in the carotid body.

Full text

Page 1

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body




Combined oxygen and glucose sensing in the
carotid body.

R PARDAL and J. LÓPEZ-BARNEO
Laboratorio de Investigaciones Biomédicas. Departamento de Fisiología and Hospital Universitario Virgen del Rocío,
Universidad de Sevilla, E-41013, Seville, Spain

INTRODUCTION
Oxygen sensing is essential for the adaptation of living organisms to variable
habitats and physiologic situations. In mammals, survival in acute hypoxia requires fast
respiratory and cardiocirculatory adjustments to guarantee proper O2 supply to the most
critical organs, such as the brain and heart (1). Reduction of arterial O2 tension is detected
by the carotid bodies, small ovoid organs situated in the carotid bifurcation, which
contain afferent nerve fibers that activate the brainstem respiratory centers to produce
hyperventilation. The O2-sensitive elements within the carotid body are the ectodermal-
derived glomus cells. These neurosecretory cells are electrically excitable (2, 3) and have
O2-sensitive potassium channels in their membranes (3, 4, 5). It is broadly accepted that
inhibition of these channels by low pO2 is a key step leading to membrane depolarization,
external calcium influx and the activation of neurotransmitter release, which, in turn,
stimulates the afferent sensory fibers. This model of chemotransduction, suggested by
electrophysiological experiments, has been confirmed by monitoring cytosolic [Ca2+] and
quantal catecholamine secretion in single cells (6-11).
The mechanism of acute O2 sensing based on the regulation of membrane
potassium channels has been demonstrated to operate in other neurosecretory systems,
such as cells in the neuroepithelial bodies of the lung (12), chromaffin cells of the adrenal
medulla (13), or PC-12 cells (14). However, some investigators have argued that the O2-
sensitive membrane electrical events are not directly implicated in the chemotransduction
because in their whole-carotid body preparations, the application of K+ channel blockers
do not increase the action potential firing frequency or secretory activity (15-17). It was
also reported that tetraethylammonium (TEA), a blocker of the O2-sensitive K+ current in
the carotid body, is unable to induce depolarization on dispersed rat glomus cells (18).
Given the discrepancies among observations in the carotid body reported by
different authors, we have developed a slice preparation of the organ to study the O2
sensitivity of glomus cells in the best possible physiological conditions (11, 19). The
Copyright © 2004 Undersea and Hyperbaric Medical Society, Inc. 113

Page 2

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
carotid body thin slice has been found to be an excellent preparation to study the cellular
bases for the chemosensitivity of glomus cells. We have found that carotid body cells are
multimodal sensor organs that detect not only the reduction of the pO2 but also the
decrease of the extracellular glucose concentration.

RESULTS AND DISCUSSION

Electrophysiological Recording and the Responses of Glomus Cells to
Hypoxia.
The procedures followed to make carotid body slices are described in detail
elsewhere (11). Figure 1A shows the aspect of one glomerulus of glomus cells within a
slice, with the arrow pointing to a well-identifiable cell, susceptible to being analyzed
electrophysiologically. The distribution of the cells in glomeruli within the slice is
consistent with the structure observed in the organ after sectioning and immunostaining
for tyrosine hydroxilase (TH), the rate limiting enzyme in the synthesis of dopamine, a
common catecholamine found in glomus cells (Figure 1B).

Figure 1. Electrophysiological recording on glomus cells within rat carotid body
thin slices.

Fig. 1 A & B. Electrophysiological
recording on glomus cells within rat
carotid body thin slices. (A) Typical
glomerulus in a slice with well-defined
single glomus cells (arrow). (B)
Carotid body slice immunostained with
antibodies against
hydroxilase. Note
appearance of glomus cells with large
nuclei and a thin layer of stained
cytoplasm.
tyrosine
typical the


Stable recordings of membrane currents can easily be obtained from glomus cells
in the slices by using the whole-cell configuration of the patch-clamp technique, as
adapted in our laboratory (20, 21). The majority of the cells recorded in the slices (77%
from a total number of 110) had inward and outward currents qualitatively similar to
those described in enzymatically dispersed rat glomus, or type I, cells (22, 23). The
remaining 23% of cells (n=26) had no measurable currents or presented a small transient
outward current, thus suggesting that they were type II, or sustentacular cells, present in
the carotid body (2, 20). The outward currents of glomus cells were highly sensitive to
the application of external TEA (5 mM caused an inhibition of 67.5±6% at +20 mV,
mean±standard deviation, n=4 cells) or the selective calcium-dependent K+ channel
blocker iberiotoxin (IbTX, 200 nM, caused an inhibition of 65% and 45% at +20 mV in
two cells tested). Therefore, as described in isolated cells (22, 23), a large proportion of
the outward current was due primarily to the activity of maxi-K+ voltage- and calcium-
dependent channels. As described in dispersed rabbit (3, 24-26) and rat (4, 22, 23, 27)

114

Page 3

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
glomus cells, the amplitude of macroscopic voltage-dependent currents was reduced
when exposed to low pO2.
Typical recordings of O2-sensitive currents are shown in Figure 1C. However, our
work on slices has focused so far on the study of secretion rather than on the recording of
ionic currents.

Fig 1C. Diagram of the whole-cell
configuration of the patch clamp
technique and
macroscopic K+ currents from a glomus
cell elicited by depolarizing pulses from
–80 mV to the indicated voltage in the
three experimental conditions (control,
low pO2, and recovery).


We have found the slice
preparation to be particularly
convenient to study
responsiveness of intact glomus
superimposed
the
cells to hypoxia and other stimuli using the amperometric detection of catecholamines (7,
10, 11). Quantal transmitter release can be monitored with a polarized 8-12-µm carbon-
fiber electrode positioned near the surface of a glomus cell (Figure 1D). The fiber is
connected to a high-gain current-to-voltage converter and polarized to +750 mV, a value
more positive than the redox potential of dopamine, the most abundant catecholamine in
glomus cells. A representative secretory response of a glomus cell to depolarization by
either a hypoxic solution or a high external K+ solution is shown in Figure 1D.

Fig. 1 D. Diagram of the amperometric detection of
catecholamine release from a single glomus cell,
and amperometric recording from an O2-sensitive
glomus cell illustrating the increase of secretory
activity induced by hypoxia and high extracellular
potassium. (Modified from Ref. 11 and 19.)

Spike-like signals resulted from the
fusion of single vesicles. The area under
individual spikes (quantal charge) yields an
estimate of the number of catecholamine
molecules released, assuming that two
electron charges are transferred to the fiber
during the oxidation of each catecholamine molecule. Average quantal charge estimated
from high K+-induced events was 40.3±17.6 fC (n=293 spikes in 7 cells), corresponding
to ≈125,000±50,000 molecules per vesicle. These values, obtained from cells in slices,
are comparable to those previously described in dispersed rat and rabbit glomus cells (7,
10, 11). The magnitude of the secretory responses of the cells was estimated either from
the number of spikes in the minute after ninety seconds of exposure to the stimulus
(frequency in events/min) or from the sum of all quantal charges measured in the same
time period and expressed as femtocoulombs per minute (fC/min; secretion rate). This

115

Page 4

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
last variable represents the total amount of catecholamine molecules released per minute
at the peak of the response.
In slices with well-defined glomeruli, low pO2 consistently induced a progressive
increase in the frequency of secretory events (Fig. 1D), from almost rest in normoxia to a
value of 48.6±19 spikes/min (n=24 cells) and a secretion rate of 1,710±270 fC/min (n=17
cells). At the peak of the response to hypoxia, the secretory events fused into a broad
concentration envelope that quickly declined after switching to the control, normoxic
solution . All the glomus cells that responded to hypoxia were also activated by solutions
with high external K+ (Fig. 1D), as expected from electrically excitable cells.
Interestingly, we also observed glomus cells that were unresponsive to hypoxia but
activated by depolarization with high external K+. Cells insensitive to hypoxia were more
frequently observed in slices that appeared somewhat unhealthy, possibly due to damage
during the experimental protocol. One possible explanation is that these are cells with O2
sensors uncoupled from the membrane ion channels. In all cells tested (n=10), the
neurosecretory response to hypoxia was completely abolished by the addition of the
voltage-dependent calcium channel blocker cadmium (Figure 2A) or the removal of
extracellular calcium with EGTA (11). This observation confirmed the dependence of the
response to hypoxia on the extracellular calcium influx through voltage-gated calcium
channels, as previously shown on dispersed rabbit carotid body cells (7, 9).

Figure 2. Secretory responses of glomus cells in the slices to cadmium and K+
channel blockers.

Fig 2A. Secretory activity recorded from
an O2-sensitive glomus cell to illustrate the
reversible abolishment of the response to
hypoxia during the blockade of Ca2+
channels by addition of 0.2 mM cadmium
to the extracellular solution.



Secretory Responses of Glomus Cells to Potassium Channel Blockers
The major contributors to the O2-sensitive macroscopic K+ currents in rat glomus
cells are voltage- and Ca2+-dependent maxi-K+ channels (4, 22, 23). Because TEA or
iberiotoxin (IbTX) block these channels, we have studied whether, like hypoxia , addition
of these agents to the external solution induces Ca2+ entry and secretion from glomus
cells. In most cells studied (33 of 34), application of 5 mM TEA to the bath solution
elicited an increase in the secretory activity similar to that triggered by hypoxia (Figure
2B).

Fig 2 B. Amperometric recording from
a glomus cell showing the similar
effects elicited by low pO2 and the
application of 5 mM TEA.

The response to the
blocker reached a frequency of

116

Page 5

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
42±17 spikes/min (n=6 cells), with a secretion rate of 1,878±470 fC/min (n=6). These
values are not significantly different from the respective ones obtained in low pO2
(Student's t-test, p=0.05). The average quantal charge of events induced by TEA was
43±30 fC (n=275 spikes in 6 cells). This value is also similar to that estimated with
events elicited by hypoxia (43±26 fC; n=576 spikes in 14 cells), suggesting that both
stimuli trigger the release of vesicles from the same cellular pool. The effect of TEA was
observed even in quiescent cells, without any measurable spontaneous quantal release, as
well as in O2-insensitive glomus cells. We have also tested the effect of IbTX, a selective
blocker of Ca2+- and voltage-activated maxi K+ channels (28). Figure 2C illustrates the
increase of secretory activity in a glomus cell exposed to 200 nM IbTX.


Figure 2 C. Secretory activity induced in a
glomus cell by hypoxia and 200 nM IbTX
(Modified from Ref. 11).


The response is similar to
those obtained with TEA or hypoxia, although the recovery phase seems to be somewhat
longer, possibly due to slower washout of IbTX. All these observations indicate that
direct blockade of the O2-sensitive K+ channels with TEA or IbTX can elicit secretion
from rat glomus cells in the slices. Our data make it difficult to understand why K+
channel blockers do not activate whole carotid body preparations, in which the glomus
cell-afferent fiber synapses are maintained intact (16, 17). As suggested before (11), a
possible explanation is that the blockers do not diffuse at the appropriate concentration
into the extracellular space of the carotid bodies either superfused by the bath solution or
perfused through the carotid artery.

Sensitivity of Glomus Cells in the Slices to Hypoglycemia
It has been proposed that the carotid body participates in glucose homeostasis (29,
30), and recently, it has been shown that resection of the carotid bodies and surrounding
tissues results in the impairment of the insulin-induced counter-regulatory responses to
mild hypoglycemia (31). However, there is no evidence
that any of the cellular elements in the carotid body can
directly respond to changes in extracellular glucose
concentration. Using the carotid body thin slice
preparation, we studied the effect of external glucose
removal on the secretory activity of the glomus cells (32).
Figure 3 represents the response of intact glomus cells to
low glucose.

Fig 3 A. Top. Amperometric signal illustrating the increase of
secretory activity in a glomus cell exposed to glucose-free solution.
Bottom. Cumulative secretion signal (in femtocoulombs) resulting
from the time integral of the amperometric recording.





117

Page 6

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
In healthy preparations, exposure of a glomus cell in the slice to 0 glucose
consistently produced a marked and reversible increase of cell secretory activity (Figure
3A). The average rate of secretion during the last minute of exposure to low glucose
(1870±386 fC/min, n=14 cells) was over twenty times (Figure 3B) that of the control
condition (88±45 fC/min, n=14).

Fig 3 B and C. (B) Bar diagram
quantifying the average response
of glomus cells to 0 glucose. (C)
Histogram representing
distribution of
exocytotic events in low glucose.

The size distribution and
mean area of quantal events
triggered by low glucose
(Figure 3C) were similar to
those previously observed in
glomus cells activated by
the
of the area
high K+, hypoxia or TEA (see above), further suggesting that all these stimuli induce the
release of a common vesicle pool. The effect of low glucose on glomus cells was
concentration-dependent (Figure 3D) and additive with the effects of hypoxia.

Fig. 3 D Secretory response of a glomus cell
to different concentrations of glucose in the
external solution.

At normal air O2 tension (pO2
≈150 mmHg), secretion was evoked
only when glucose decreased below
2 mM. However, at a pO2 of 90
mmHg (a value close to the normal
O2 tension in arterial blood), glomus
cell secretory
significantly modulated by glucose
activity was
in the concentration range (2 to 5 mM) that includes the values observed in common
hypoglycemic situations (32; Figure 3E).



Fig. 3E. Logarithm of maximal secretion rate
(ordinate) versus glucose concentration at two
different pO2 values. (Modified from Ref. 23.)


Catecholamine secretion induced by
glucose-free solutions
suppressed by the addition of 0.2 mM Cd2+
to the extracellular solution in all cells
was totally

118

Page 7

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
tested (n=3; Figure 4A). Because 0.2 mM extracellular Cd2+ completely blocks voltage-
gated Ca2+ channels in glomus cells (7), this indicates that low glucose induces
transmitter secretion when membrane electrical events result in depolarization and Ca2+
influx through voltage-dependent channels. Experiments on patch-clamped cells dialyzed
with an internal solution containing 4 mM Mg-ATP confirmed this idea. Glucose
deficiency produced a reversible reduction of peak outward K+ current amplitude
(Figures 4B and 4C) that at +20 mV, had an average value of 38±12% (n=7 cells). Low
glucose appeared to act selectively on voltage-dependent K+ channels since it had no
effect on the small inward current characteristic of most rat glomus cells (data not
shown).

Figure 4. Glomus cell membrane electrical events in the response to low glucose.


Fig. 4. Glomus cell membrane
electrical events in the response to
low glucose.
suppression of low glucose-evoked
secretory activity by application of
0.2 mM cadmium
extracellular solution.
Recordings of outward K+ currents
from a patch-clamped glomus cell
depolarized to 0 and +20 mV and
exposed to 0 mM glucose. The
control (c) and recovery (r)
external solutions contained 5 mM
glucose. (C)
relationship obtained measuring
maximum current amplitudes from
the experiment shown on B.

The electrophysiological
and amperometric
obtained from glomus cells
in carotid
(A) Reversible
to the
(B)
Current-voltage
data
body slices
strongly suggest that these cells are physiological low-glucose detectors capable of
transducing glucose levels into variable rates of transmitter release. The low-glucose
signaling pathways in glomus cells appear to be initiated by an inhibition of voltage-
gated K+ channel activity, which leads to membrane depolarization, Ca2+ influx through
voltage-gated Ca2+ channels and transmitter release. Therefore, in glomus cells, low
glucose and hypoxia converge to raise cytosolic [Ca2+] and release transmitters, which
stimulates afferent sensory fibers and evokes sympathoadrenal activation. These
observations help explain previous reports of anesthetized animals exhibiting rapid
increases in the output of hepatic glucose after the activation of the carotid body with
sodium cyanide (29, 30), alterations of carbohydrate metabolism in acute hypoxia (33), or
the impairment of an insulin-induced counter-regulatory response to mild hypoglycemia
in carotid body resected dogs (31). Although the existence of peripheral glucosensors,
presumably located in the liver or portal vein, has been proposed (34, 35), the

119

Page 8

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body
strategically located carotid bodies may be of special importance for brain homeostasis,
as neurons are particularly vulnerable to the simultaneous lack of glucose and oxygen
(36). The function of glomus cells as combined O2 and glucose sensors, in which the two
stimuli potentiate each other, is surely advantageous to facilitate the activation of
counter-regulatory measures in response to small reductions of any of the regulated
variables. Impairment of low-glucose sensing by carotid body glomus cells might
contribute to the susceptibility of insulin-dependent diabetic patients to hypoglycemia.

REFERENCES

1. López-Barneo J, Pardal R and Ortega-Sáenz P. Cellular mechanisms of oxygen sensing. Annu Rev
Physiol 2001; 63: 259-287.
2. Duchen MR, Caddy KWT, Kirby GC, Patterson DL, Ponte J and Biscoe TJ. Biophysical studies of the
cellular elements of the rabbit carotid body. Neuroscience 1988; 26: 291-311.
3. López-Barneo J, López-López JR, Ureña J and González C. Chemotransduction in the carotid body: K+
current modulated by PO2 in type I chemoreceptor cells. Science 1988; 241:580-582.
4. Peers C. Hypoxic suppression of K+ currents in type I carotid body cells: selective effect on the Ca2+-
activated K+ current. Neurosci Lett 1990; 119:253-256.
5. Buckler KJ. A novel oxygen-sensitive potassium current in rat carotid body type I cells. J Physiol 1997;
498:649-662.
6. López-Barneo J, Benot AR and Ureña, J. Oxygen sensing and the electrophysiology of arterial
chemoreceptor cells. News Physiol Sci 1993; 8: 191-195.
7. Ureña J, Fernández-Chacón R, Benot AR, Álvarez de Toledo G and López-Barneo J. Hypoxia induces
voltage-dependent Ca2+ entry and quantal dopamine secretion in carotid body glomus cells. Proc Natl
Acad Sci USA 1994; 91:10208-10211.
8. Buckler KJ and Vaughan-Jones RD. Effects of hypoxia on membrane potential and intracellular calcium
in rat neonatal carotid body type I cells. J Physiol 1994; 476:423-428.
9. Montoro RJ, Ureña J, Fernández-Chacón R, Álvarez de Toledo G and López-Barneo J. Oxygen sensing
by ion channels and chemotransduction in single glomus cells. J Gen Physiol 1996; 107:133-143.
10.Carpenter E, Hatton CJ and Peers C. Effects of hypoxia and dithionite on catecholamine release from
isolated type I cells of the rat carotid body. J Physiol 2000; 523:719-729.
11. Pardal R, Ludewig U, García-Hirschfeld J and López-Barneo J. Secretory responses of intact glomus
cells in thin slices of rat carotid body to hypoxia and tetraethylammonium. Proc Natl Acad Sci USA
2000; 97:2361-2366.
12. Youngson C, Nurse C, Yeger H and Cutz E. Oxygen sensing in airway chemoreceptors. Nature 1993;
365:153-155.
13. Thompson RJ and Nurse CA. Anoxia differentially modulates multiple K+ currents and depolarizes
neonatal rat adrenal chromaffin cells. J Physiol 1998; 512: 421-434.
14. Zhu WH, Conforti L, Czyzyk-Krzeska MF and Millhorn DE. Membrane depolarization in PC12 cells
during hypoxia is regulated by an O2-sensitive K+ current. Am J Physiol 1996; 271:C658-C665.
15. Doyle TP and Donnelly DF. Effect of Na+ and K+ channel blockade on baseline and anoxia induced
catecholamine release from rat carotid body. J Appl Physiol 1994; 77: 2606-2611.
16. Osanai S, Buerk DG, Mokashi A, Chugh DK and Lahiri S. Cat carotid body chemosensory discharge
(in vitro) is insensitive to charybdotoxin. Brain Res 1997; 747: 324-327.
17. Lahiri S, Roy A, Rozanov C and Mokashi A. K+-current modulated by PO2 in type I cells in rat carotid
body is not a chemosensor. Brain Res 1998; 794: 162-165.
18. Buckler KJ. A novel oxygen-sensitive potassium current in rat carotid body type I cells. J Physiol 1997;
498:649-662.
19. Pardal R and López-Barneo J. Carotid body thin slices: responses of glomus cells to hypoxia and K+
channel blockers. Respir Physiol Neurobiol 2002; 132:69-79.
20. Ureña J, López-López JR, González C and López-Barneo J. Ionic currents in dispersed chemoreceptor
cells of the mammalian carotid body. J Gen Physiol 1989; 93: 979-999.
21. Ganfornina MD and López-Barneo J. Potassium channel types in arterial chemoreceptor cells and their
selective modulation by oxygen. J Gen Physiol 1992; 100: 401-426.

120

Page 9

UHM 2004, Vol. 31, No. 1 – Glucose and O2 sensing by the carotid body

121
22. Wyatt CN and Peers C. Ca2+-activated K+ channels in isolated type I cells of the neonatal rat carotid
body. J Physiol 1995; 483: 559-565.
23. López-López JR, González C and Pérez-García MT. Properties of ionic currents from isolated adult rat
carotid body chemoreceptor cells: effect of hypoxia. J Physiol 1997; 499: 429-441.
24. Delpiano MA and Hescheler J. Evidence for a PO2-sensitive K+ channel in the type-I cell of the rabbit
carotid body. FEBS Lett 1989; 249:195-198.
25.Ganfornina MD and López-Barneo J. Single K+ channels in membrane patches of arterial
chemoreceptor cells are modulated by O2 tension. Proc Natl Acad Sci USA 1991; 88:2927-2930.
26. Pérez-García T, López-López JR, Riesco AM, Hoppe UC, Marbán E, González C and Johns DC. Viral
gene transfer of dominant negative Kv4 construct suppresses an O2 sensitive K+ current in
chemoreceptor cells. J Neurosci 2000; 20: 5689-5695.
27. Stea A and Nurse CA. Whole-cell and perforated-patch recordings from O2-sensitive rat carotid body
cells grown in short- and long-term culture. Pflügers Arch 1991; 418:93-101.
28. Gálvez A, Gimenez-Gallego G, Reuben JP, Roy-Contancin L, Feigenbaum P, Kaczorowski GJ and
García ML. Purification and characterization of a unique, potent, peptidyl probe for the high
conductance calcium-activated potassium channel from venom of the scorpion Buthus tamulus. J Biol
Chem 1990; 265: 11083-11090.
29. Álvarez-Buylla R and Álvarez-Buylla ER. Carotid sinus receptors participate in glucose homeostasis.
Respir Physiol 1988; 72: 347-360.
30. Álvarez-Buylla R and Álvarez-Buylla ER. Changes in blood glucose concentration in the carotid body
sinus modify brain glucose retention. Brain Res 1994; 654: 167-170.
31. Koyama Y, Coker RH, Stone EE, Lacy DB, Jabbour K, Williams PE and Wasserman DH. Evidence
that carotid bodies play an important role in glucoregulation in vivo. Diabetes 2000; 49: 1434-1442.
32. Pardal R and López-Barneo J. Identification of low-glucose sensing cells in the carotid body. Nat
Neurosci 2002; 5:197-198.
33. Zinker BA, Namdaran K., Wilson R, Lacy DB and Wasserman DH. Acute adaptation of carbohydrate
metabolism to decreased arterial PO2. Am J Physiol 1994; 266: E921-E929.
34. Donovan CM, Hamilton-Wessler M, Alter JB and Bergman RN. Primacy of liver glucosensors in the
sympathetic response to progressive hypoglycemia. Proc Nat Acad Sci USA 1994; 91: 2863-2867.
35. Hevener AL, Bergman RN and Donovan CM. Novel glucosensor for hypoglycemia detection localized
to the portal vein. Diabetes 1997; 46: 1521-1525.
36. Martin RL, Lloyd HG and Cowan AI. The early events of oxygen and glucose deprivation: setting the
scene for neuronal death? Trends Neurosci 1994; 17:251-257.